From c721472fb83d1f7c7606bcf33ba2d42d6127a764 Mon Sep 17 00:00:00 2001
From: "Miss Islington (bot)"
 <31488909+miss-islington@users.noreply.github.com>
Date: Fri, 13 Jul 2018 20:58:12 -0700
Subject: bpo-34087: Fix buffer overflow in int(s) and similar functions
 (GH-8274)

`_PyUnicode_TransformDecimalAndSpaceToASCII()` missed trailing NUL char.
It caused buffer overflow in `_Py_string_to_number_with_underscores()`.

This bug is introduced in 9b6c60cb.
(cherry picked from commit 16dfca4d829e45f36e71bf43f83226659ce49315)

Co-authored-by: INADA Naoki <methane@users.noreply.github.com>
---
 Lib/test/test_complex.py                                              | 3 +++
 Lib/test/test_float.py                                                | 3 +++
 Lib/test/test_long.py                                                 | 4 ++++
 .../next/Core and Builtins/2018-07-13-22-09-55.bpo-34087.I1Bxfc.rst   | 1 +
 Objects/unicodeobject.c                                               | 2 ++
 Python/pystrtod.c                                                     | 2 ++
 6 files changed, 15 insertions(+)
 create mode 100644 Misc/NEWS.d/next/Core and Builtins/2018-07-13-22-09-55.bpo-34087.I1Bxfc.rst

diff --git a/Lib/test/test_complex.py b/Lib/test/test_complex.py
index 2d883c5..21c6eae 100644
--- a/Lib/test/test_complex.py
+++ b/Lib/test/test_complex.py
@@ -345,6 +345,9 @@ class ComplexTest(unittest.TestCase):
         self.assertEqual(type(complex("1"*500)), complex)
         # check whitespace processing
         self.assertEqual(complex('\N{EM SPACE}(\N{EN SPACE}1+1j ) '), 1+1j)
+        # Invalid unicode string
+        # See bpo-34087
+        self.assertRaises(ValueError, complex, '\u3053\u3093\u306b\u3061\u306f')
 
         class EvilExc(Exception):
             pass
diff --git a/Lib/test/test_float.py b/Lib/test/test_float.py
index 17174dd..06ea90c 100644
--- a/Lib/test/test_float.py
+++ b/Lib/test/test_float.py
@@ -60,6 +60,9 @@ class GeneralFloatCases(unittest.TestCase):
         # extra long strings should not be a problem
         float(b'.' + b'1'*1000)
         float('.' + '1'*1000)
+        # Invalid unicode string
+        # See bpo-34087
+        self.assertRaises(ValueError, float, '\u3053\u3093\u306b\u3061\u306f')
 
     def test_underscores(self):
         for lit in VALID_UNDERSCORE_LITERALS:
diff --git a/Lib/test/test_long.py b/Lib/test/test_long.py
index cc48259..8472889 100644
--- a/Lib/test/test_long.py
+++ b/Lib/test/test_long.py
@@ -373,6 +373,10 @@ class LongTest(unittest.TestCase):
         for base in invalid_bases:
             self.assertRaises(ValueError, int, '42', base)
 
+        # Invalid unicode string
+        # See bpo-34087
+        self.assertRaises(ValueError, int, '\u3053\u3093\u306b\u3061\u306f')
+
 
     def test_conversion(self):
 
diff --git a/Misc/NEWS.d/next/Core and Builtins/2018-07-13-22-09-55.bpo-34087.I1Bxfc.rst b/Misc/NEWS.d/next/Core and Builtins/2018-07-13-22-09-55.bpo-34087.I1Bxfc.rst
new file mode 100644
index 0000000..5147395
--- /dev/null
+++ b/Misc/NEWS.d/next/Core and Builtins/2018-07-13-22-09-55.bpo-34087.I1Bxfc.rst	
@@ -0,0 +1 @@
+Fix buffer overflow while converting unicode to numeric values.
diff --git a/Objects/unicodeobject.c b/Objects/unicodeobject.c
index d5e7d10..5d605ab 100644
--- a/Objects/unicodeobject.c
+++ b/Objects/unicodeobject.c
@@ -9076,6 +9076,7 @@ _PyUnicode_TransformDecimalAndSpaceToASCII(PyObject *unicode)
             int decimal = Py_UNICODE_TODECIMAL(ch);
             if (decimal < 0) {
                 out[i] = '?';
+                out[i+1] = '\0';
                 _PyUnicode_LENGTH(result) = i + 1;
                 break;
             }
@@ -9083,6 +9084,7 @@ _PyUnicode_TransformDecimalAndSpaceToASCII(PyObject *unicode)
         }
     }
 
+    assert(_PyUnicode_CheckConsistency(result, 1));
     return result;
 }
 
diff --git a/Python/pystrtod.c b/Python/pystrtod.c
index 9bf9363..141a47a 100644
--- a/Python/pystrtod.c
+++ b/Python/pystrtod.c
@@ -391,6 +391,8 @@ _Py_string_to_number_with_underscores(
     char *dup, *end;
     PyObject *result;
 
+    assert(s[orig_len] == '\0');
+
     if (strchr(s, '_') == NULL) {
         return innerfunc(s, orig_len, arg);
     }
-- 
cgit v0.12